RESEARCH

Dyspepsia prevalence and impact on quality of life among

Rwandan healthcare workers: A cross-sectional survey
R Bitwayiki,1,2 MBBS, MMed (IM); J T Orikiiriza,3,4,5 MB ChB, MMed (Paed); F Kateera,3 MB ChB, MSc; P Bihizimana,1,2 MBBS, MMed (IM);
B Karenzi,3 DBA, MBA, MA; P Kyamanywa,2 MB ChB, FACS; T D Walker,1,2 MBBS, FRACP, MPHTM
University Teaching Hospital, Butare, Rwanda
School of Medicine, College of Medicine and Health Sciences, University of Rwanda, Butare, Rwanda
3
Rwanda Military Hospital, Kigali, Rwanda
4
Infectious Diseases Institute, School of Medicine and Health Sciences, Makerere University, Kampala, Uganda
5
Department of Immunology, Trinity College, Dublin, Ireland
1

2

Corresponding author: T D Walker (timwalkerd@gmail.com)

Background. Dyspepsia has been demonstrated worldwide to have major personal and societal impacts, but data on the burden of this
disease in Africa are lacking.
Objective. To document the prevalence of dyspepsia and its quality-of-life impact among healthcare workers (HCWs) at Butare University
Teaching Hospital (BUTH), Rwanda.
Methods. A cross-sectional survey among consenting HCWs at BUTH was conducted. Multilingual interviewers guided participants
through validated questionnaires, including the Short-Form Leeds Dyspepsia Questionnaire (SF-LDQ), to detect the presence and
frequency of dyspeptic symptoms, and the Short-Form Nepean Dyspepsia Index (SF-NDI), to examine the impact of dyspepsia on quality
of life.
Results. The study included 378 enrolled HCWs, all of whom provided responses to the SF-LDQ and 356 of whom responded to the
SF-NDI. The prevalence of dyspepsia in the study population was 38.9% (147/378). Of these 147 HCWs, 79 (53.7%) had very mild
dyspepsia, 33 (22.4%) had mild dyspepsia, 20 (13.6%) had moderate dyspepsia and 15 (10.2%) had severe dyspepsia. Females were
more likely to complain of dyspepsia than males (98/206 v. 49/172; odds ratio (OR) 2.3; 95% confidence interval (CI) 1.5 - 3.5; p<0.001).
Participants with dyspepsia of at least mild severity had SF-NDI scores reflecting reduced quality of life when compared with non-dyspeptic
participants (OR 17.0; 95% CI 5.0 - 57.1; p<0.001), with most marked effects on the tension and eating and drinking subdomains of the
SF-NDI.
Conclusion. The prevalence of dyspepsia among HCWs in Rwanda is high and is associated with lowered quality of life.
S Afr Med J 2015;105(12):1064-1069. DOI:10.7196/SAMJ.2015.v105i12.9482

Dyspepsia is a common presenting symptom com

plex in primary care all over the world, accounting
for 3 - 4% of primary healthcare visits.[1] Populationbased surveys in many countries have revealed a
highly variable prevalence, from 1.8% to 57%, with
a global average of 20.8% in the published literature.[2] These
geographical differences are not fully accounted for by variations in
symptom definition, although, as expected, large differences have
been noted between less strict and more strict definitions of the dys
peptic symptom complex.[2]
Data on the frequency of dyspepsia in primary care settings in
Africa are lacking, with only two prior studies having been published,
both from Nigeria.[3,4] However, while dyspepsia is a frequent reason
for clinic and hospital attendance across Africa, it is not clear what
proportion of symptomatic patients seek medical treatment. No
previous studies have looked at the impact on quality of life of
dyspeptic symptoms in Africans living in the community, although
studies in other populations have shown a significant impact on
quality of life, but not mortality.[5,6]
The vast majority of African studies of dyspepsia have been done in
endoscopy or tertiary clinic referral populations. These populations
are not representative of the community, for whom the burden of
disease is unknown. This study seeks to reduce this knowledge gap
by characterising the burden of dyspepsia in a presumed well, clearly
defined sample of healthcare workers (HCWs) in Rwanda, to provide

1064

an accurate measure of the frequency and distribution of dyspeptic

symptoms among a community-representative sample, and the
quality-of-life costs associated with their symptoms.

Methods

This cross-sectional survey was conducted in Butare University

Teaching Hospital (BUTH), located in the Huye District in the
Southern Province of Rwanda, during October and November 2013,
in conjunction with a larger study evaluating viral hepatitis burden
and risk determinants among HCWs. The hospital is the sole tertiary
referral centre for Southern Rwanda, with 500 beds and a catchment
population of about 5 million people. The hospital also serves as a
teaching site for Rwandas sole medical school. The hospital employed
747 personnel at the time of the study: 419 clinical HCWs, 78
administrative HCWs and 250 cleaners.

Study site and sample size

The survey was powered to assess the prevalence of dyspepsia in the

population. The surveys sample size was estimated at 373 subjects,
based on a closed eligible population of 747 HCWs at BUTH, with a
presumed dyspepsia prevalence of 35% based on prior communitybased African studies,[3,4] and aiming for a 5% relative standard error
in the measurement of dyspepsia prevalence at an alpha level of
0.05. All HCWs were invited to attend the study centre for free viral
hepatitis testing, and then were offered participation in this study

December 2015, Vol. 105, No. 12

RESEARCH

on their arrival, independent of the testing. Sampling of all eligible

HCWs was consecutive, without random selection.

Inclusion and exclusion criteria

All consenting HCWs aged 18 years and currently employed at

BUTH were eligible for study enrolment. HCWs on leave or not
currently working because of illness at the time of the study were
not invited. Postgraduate students on clinical placements were
considered as HCWs, but undergraduate students were not eligible.

Questionnaires

Questionnaires were administered and filled out by trained

interviewers with language skills in English, French and Kinyarwanda.
The questionnaire was made up of three sections: (i) socio
demographic data; (ii) the Short-Form Leeds Dyspepsia Questionnaire
(SF-LDQ); and (iii) the Short-Form Nepean Dyspepsia Index
(SF-NDI).
The SF-LDQ was used to detect the presence and frequency of
dyspeptic symptoms, and the SF-NDI to examine the impact of
dyspepsia on quality of life.
Where possible, the survey was administered in English, but if
necessary the interviewer translated the questions into French or
Kinyarwanda. Prior to survey administration, an agreed wording
for the oral translation of the English instruments into French
and Kinyarwanda was completed, involving translation and backtranslation of the survey instrument by a group of multilingual
medical experts. The LDQ and SF-NDI instruments were tested for
reliability in a pilot study comprising ten HCWs eligible for study
entry but not forming part of the final data set. Qualitative data were
obtained from this focus group to improve the questionnaire.
The Short-Form Leeds Dyspepsia Questionnaire
The SF-LDQ consists of eight items that assess for the presence
and severity of dyspepsia by measuring the frequency and severity
of upper abdominal pain/discomfort, heartburn, regurgitation and
nausea. It was developed from the Leeds Dyspepsia Questionnaire
(LDQ), a longer form survey.[7] Possible scores range from 0 to
32 with higher values corresponding with increasing severity of
dyspepsia. The developers of the SF-LDQ have defined a score of 0
as no dyspepsia, a score of 1 - 4 as very mild dyspepsia, a score of
5-8 as mild dyspepsia, a score of 9-15 as moderate dyspepsia
and a score >15 as indicative of severe dyspepsia.[8] In the design
of the study, we planned to analyse these results categorically but
also dichotomously, with those in the no dyspepsia and very mild
dyspepsia groups defined as non-dyspeptic (i.e. those with an
SF-LDQ score of <5). The LDQ and SF-LDQ have been used in many
studies and have previously been validated as accurate and reliable
tools to assess the presence and severity of dyspepsia in multiple,
diverse and multicultural populations.[9,10]
The Short-Form Nepean Dyspepsia Index
The SF-NDI is a questionnaire comprising ten items, making up five
subscales of two items each that examine the impact of dyspepsia on
various domains of the quality of life of patients, including tension/
anxiety, disruption of regular eating/drinking, knowledge and control
over disease symptoms and interference with work/study.[11] It is also
based on an earlier tool, the Nepean Dyspepsia Index (NDI), which
was developed to study quality of life in functional dyspepsia.[12]
Each item is measured by a 5-point Likert scale ranging from 1
(not at all or not applicable) to 5 (extremely). Individual items in
each subscale are added to obtain a score range from 10 (lowest score)
to 100 (highest score). A higher score on the SF-NDI is indicative of

1065

a poorer quality of life, with patients with scores >15 having been
shown by the designers of this instrument to have significantly
reduced health-related quality of life. The NDI and SF-NDI have been
used in many studies and have been validated in various populations
as accurate and reliable tools to measure the impact of dyspepsia on
quality of life.[13,14]

Data analysis

Key outcomes for this study include the frequency and quality-oflife impact of dyspepsia, measured by the SF-LDQ and SF-NDI,
respectively. Demographic data (age, gender, marital status,
department of work, profession and educational level), smoking, and
alcohol, aspirin and non-steroidal drug intake are probable predictors
of dyspepsia, while we expect these factors may also be co-variable
with the quality-of-life impact of dyspepsia, and these were assessed
in the sociodemographic section of the questionnaire.
All data were collected using a questionnaire uploaded on personal
digital assistants. Query programs were written into the database to
limit entry of incorrect data and to ensure data quality. The study
co-ordinator and data manager reviewed the data and all associated
queries daily for completeness and accuracy. Any missing data were
coded as such and excluded from analysis of that variable.
The data were then transferred into SPSS (version 16.0; IBM, USA)
and Stata (version 12.1; Stata Corp, USA) for data analysis. Outcome
variables were the scores on SF-LDQ and SF-NDI. Comparisons
of characteristics between groups of study participants were made
using the 2 test for categorical variables and one-way analysis
of variance (ANOVA) for non-normally distributed continuous
variables. Associations between predictor variables and outcomes of
interest were estimated using both univariate and multivariate logistic
regression, for those variables found to be significant by univariate
analysis. In the final interpretation of results, a p-value of <0.05 was
considered statistically significant.

Ethical considerations

Written informed consent was required from all study participants.

The study was approved by the Butare University Teaching Hospital
Research and Ethics Commission and the University of Rwanda
School of Medicine Research Commission.

Results

The questionnaire was administered to 378 HCWs at BUTH, all of

whom provided responses to SF-LDQ and 356 of whom responded
to the SF-NDI.
The study population was generally young (mean 34.1 years,
range 18-63), with a slight female preponderance (54.5%).
The majority were tertiary educated (58.7%), but a significant
proportion (18%) had not attended secondary school. Ancillary
staff, including cleaners and administration staff, outnumbered
both nurses and doctors in the study population. All major hospital
departments were represented, in approximate proportion to their
number of staff (Table 1). The staff included in the study did not
differ significantly in age, gender or department of work from those
who were not included.
Using a broad definition of (any) dyspepsia, the prevalence of
dyspepsia in the study population was 38.9% (147/378). Of these
147 HCWs, 79 (53.7%) had very mild dyspepsia, 33 (22.4%) mild
dyspepsia, 20 (13.6%) moderate dyspepsia and 15 (10.2%) severe
dyspepsia. Females were more likely to complain of dyspepsia
than males (98/206 v. 49/172, odds ratio (OR) 2.3; 95% confidence
interval (CI) 1.5 - 3.5; p<0.001). Higher education levels were asso
ciated with a lower risk of dyspeptic symptoms (OR 0.6; p=0.015),

December 2015, Vol. 105, No. 12

RESEARCH

Table 1. Baseline characteristics and

distribution of 378 study participants

Table 2. Dyspepsia prevalence and its association with demographic and lifestyle
factors, by univariate analysis
Dyspepsia, n/total (%)

OR (CI)

Variable

n (%)

Gender

Overall

Male

172 (45.5)

Gender

Female

206 (54.5)

Male

49/172 (28.5)

(ref.)

Female

98/206 (47.6)

2.28 (1.48 - 3.50)

Age group (years)

125 (33.1)

30 - 49

232 (61.3)

Single

45/131 (34.4)

50

21 (5.6)

Married

96/232 (41.4)

Widowed

4/9 (44.4)
2/6 (33.3)

Single

131 (34.6)

Other

Married

232 (61.4)

Age (years)

Divorced/separated

4 (1.1)

<30

64/150 (42.7)

Widowed

9 (2.4)

30 - 49

69/193 (35.8)

Living together

2 (0.5)

50

14/35 (40.0)

Current department of work

NS

NS

Education level

Medicine

47 (12.4)

Primary or secondary*

72/156 (46.2)

(ref.)

Obstetrics & gynaecology

30 (7.9)

Tertiary

75/222 (33.8)

0.60 (0.39-0.91)

Paediatrics

47 (12.4)

Smoking

Surgery

52 (13.9)

Never

133/339 (39.2)

Administration

28 (7.4)

Past

10/32 (31.3)

Laboratory

11(2.9)

Current

4/7 (57.1)

Sterilisation

6 (1.7)

 ther departments (cleaning

O
services, etc.)
156 (41.4)
Education

Nil

101/241 (41.9)

Any current

46/137 (33.6)

Never

29/81 (35.8)

Primary level

65 (17.2)

Past

96/256 (37.5)

Secondary/vocational

88 (23.3)

Current

22/43 (51.2)

University/other tertiary

222 (58.7)

(ref.) = referent population for odds ratio calculations.

*NB. The primary and secondary education group includes HCWs with vocational training.

123 (32.5)

Doctors

40 (10.6)

 thers (cleaners, laboratory

O
staff, etc.)
215 (56.9)

but conventional risk factors for dyspepsia,

including smoking and use of alcohol, nonsteroidal anti-inflammatory medications
and aspirin, were not associated with
dyspeptic symptoms in this population
(Table 2). Indigestion was the most
frequent symptom type reported (in 25%
of study subjects), with heartburn (22%),
regurgitation (15%) and nausea (9%) all
both less frequently observed and showing
lower frequency of lifestyle interference on
the SF-LDQ (Table 3).
On multivariate analysis, gender
(p<0.001) and level of education (p=0.006)
remained highly predictive of dyspepsia,

NS

NS

NSAID use

3 (0.8)

Nurses

0.015

Alcohol

None

Staffing group

<0.001

Marital status

18 - 29

Marital status

p-value

147/378 (38.9)

but no other factors in the model were

statistically significant, with the exception
of the consumption of alcohol, which was
associated with a borderline lower risk of
dyspepsia (p=0.043).
Participants with dyspepsia had reduced
quality-of-life scores, as measured by
the SF-NDI, when compared with nondyspeptic participants (p<0.001), the
most marked effects being on the tension
and eating and drinking subdomains of
the SF-NDI. An appreciable difference in
quality of life was apparent, with the most
marked reduction seen in subjects with at
least mild severity dyspepsia, compared
with those suffering no or very mild
dyspepsia (sum of squares by ANOVA
779.27; p<0.001) (Table 4).
As an alternative measure, the proportion
of subjects with reduced quality of life (def

1066

December 2015, Vol. 105, No. 12

NS

ined as an SF-NDI of 15 points, consisting

of the bottom 8% of the population) in
each dyspeptic symptom group was also
calculated, with 1.4% and 10.7% of non- and
very mildly dyspeptic patients meeting this
threshold, compared with 25.8%, 31.6% and
33.3% of mildly, moderately and severely
dyspeptic patients.
Analysis of the predominant symptom and
covariance in symptoms on the SF-LDQ was
also undertaken, with the goal of defining
patients into separate dyspeptic symptom
complex groups. These were defined as
predominant epigastric pain (ulcer-like),
predominant heartburn or regurgitation
(reflux-like), predominant nausea (dys
motility-like) and mixed groups. Of 90
patients with dyspepsia and a predominant
symptom available for analysis, 38 (42.2%)
described ulcer-like symptoms, 37 (41.1%)

RESEARCH

reflux-like symptoms, 8 (8.9%) dysmotility-like symptoms, and 7

(7.8%) mixed symptoms.
There was a significantly higher SF-LDQ (t=3.314, p=0.009)
among patients with reflux-like or mixed symptoms (mean 8.25)
than among those with ulcer-like or dysmotility-like symptoms
(mean 4.94). However, quality of life measured by the SF-NDI was
not significantly different between these groups (means 13.16 v.
12.48; p>0.05).
Table 3. Frequency and lifestyle interference of each SF-LDQ
symptom type in study subjects
Symptom
frequency (%)

Interference with
lifestyle frequency (%)

Never

281 (74)

322 (85)

< monthly

34 (9)

10(3)

> monthly, < weekly

31 (8)

20 (5)

> weekly, < daily

23 (6)

17 (5)

> daily

9 (2)

9(2)

Never

298 (79)

329 (87)

< monthly

26 (7)

11 (3)

> monthly, < weekly

26 (7)

14 (4)

> weekly, < daily

18 (5)

14 (4)

> daily

10 (3)

10 (3)

Never

320 (85)

353 (93)

< monthly

23 (6)

7 (2)

> monthly, < weekly

18 (5)

8 (2)

> weekly, < daily

12 (3)

6 (2)

> daily

5 (1)

4 (1)

Never

343 (91)

355 (94)

< monthly

20 (5)

9 (2)

> monthly, < weekly

4 (1)

4 (1)

> weekly, < daily

6 (2)

5 (1)

> daily

5 (1)

5 (1)

Symptom type
Indigestion

Heartburn

Regurgitation

Nausea

Discussion

Dyspepsia affected more than a third of Rwandan HCWs in this

study, and had a significant impact on wellbeing, with symptoms
severe enough to impair the quality of life of almost a third of affected
HCWs. Ulcer-like and reflux-like symptoms were equally common,
perhaps surprisingly, given how infrequently gastro-oesophageal
reflux is diagnosed in African endoscopic series. The observed
dyspepsia prevalence is at the high end of the range garnered from
studies in other parts of the world, although these frequencies depend
greatly on the definition of dyspepsia used in the study.[2]
One of the key challenges in studying dyspepsia is a lack of a
clear, agreed standard definition of dyspeptic symptoms. Older
studies tended to use broad descriptive questions on the presence of
epigastric discomfort. More recently, the Rome II-III criteria have
been developed, which attempt to draw a distinction between reflux
symptoms and dyspepsia in order to distinguish gastro-oesophageal
reflux disease from functional dyspepsia, but have met with a mixed
reception. As a result, a variety of definitions remain in use, with
simplicity, prior multicultural validity and ease of administration
being our key criteria in choosing the SF-LDQ and SF-NDI for this
study.
The SF-LDQ is based upon a broad definition of dyspepsia, not
attempting to exclude reflux symptoms from the definition, and
assessing symptoms over a 2-month period. In a recent meta-analysis
by Ford et al.,[2] using a broad definition of this type gave a 29.5%
pooled prevalence of dyspepsia. However, very few of the studies
previously reported have used face-to-face interviews; those that
did reported higher than average prevalences, pointing to a possible
systematic effect of this data acquisition method on reported dyspepsia
prevalence. While the SF-LDQ only assesses a limited range of
dyspeptic symptoms, and does not include assessment of postprandial
symptoms, which have been highly prevalent in Western studies,
correlation between the SF-LDQ and other lengthier tools that include
such symptoms has been high in published validation studies.[8,9]
As a symptom complex, uninvestigated dyspepsia does not have
a single underlying pathological mechanism. Putative contributing
factors include Helicobacter pylori infection, gastric acid and gastric
dysmotility.[15] In Western populations, about 80% of patients with
uninvestigated dyspepsia have a final diagnosis of functional dyspepsia
following investigation.[6] The pathogenesis of functional dyspepsia is
debated, and has not been well studied in areas with a high prevalence
of H. pylori infection, although a recent observational study suggested
a role for H. pylori and associated microscopic duodenitis.[16]
Traditional risk factors such as alcohol, smoking and nonsteroidal anti-inflammatory drug (NSAID) use were not found

Table 4. Severity of dyspepsia and impact on quality of life, as measured by mean SF-NDI scores in 356 subjects*
No dyspepsia

Very mild dyspepsia

Mild dyspepsia

Moderate dyspepsia

Severe dyspepsia

p-value

SF-NDI total score

10.31

11.21

13.90

14.47

15.20

<0.001

SF-NDI: Tension

2.10

2.57

3.39

3.68

3.93

<0.001

SF-NDI: Interference with

ADLs

2.06

2.24

2.68

2.84

2.87

<0.001

SF-NDI: Eating/drinking

2.04

2.19

2.71

3.37

3.07

<0.001

SF-NDI: Knowledge/
control

2.06

2.11

2.55

2.16

2.67

0.004

SF-NDI: Work/study

2.05

2.11

2.58

2.42

2.67

<0.001

ADLs = activities of daily living.

*SF-NDI scores range from 10 to 50, with higher scores indicating more severely impaired quality of life. The last five rows refer to SF-NDI subdomains, each of which range from 2 to 10. p-values
were calculated by Mann-Whitney U-tests, comparing patients with no or very mild dyspepsia with those with more severe dyspepsia.

1067

December 2015, Vol. 105, No. 12

RESEARCH

to be associated with symptoms of dyspepsia in our study. While

this initially appears surprising and is in contrast to the wider
literature,[2] there are several good reasons why this might be
the case among HCWs. First, HCWs are more likely to have an
understanding of the link between such risk factors and the onset
of dyspeptic symptoms. Secondly and more speculatively, HCWs
might be more amenable to changing their behaviours in response
to this knowledge, even without medical attention. As this study was
cross-sectional, it was unable to examine the effects of dyspepsia on
these health-related behaviours over time.
Among Rwandan HCWs in this study, women were more likely
to complain of dyspeptic symptoms than men, and more likely
to suffer lower quality of life as a result. This is in contrast with
the only previous studies of dyspepsia in Africa, conducted in
Nigeria,[3,4] where men had a similar rate to women (OR 0.93; 95%
CI 0.73 - 1.19.[2] It should be noted that there are significant cultural
variations between African nations, and it may be that the role of
gender in illness behaviour is significantly different in Nigeria and
Rwanda, as has been noted in other illnesses in disparate African
contexts. The findings in our study are, however, quite consistent
with the broader population-based international literature, where
in a recent meta-analysis by Ford et al.[2] the pooled OR for women
with dyspepsia requiring investigation was 1.24.
Higher levels of education were protective in this study of HCWs,
although this may not be generalisable to non-healthcare education,
as most participants in this study had higher qualifications in the
health sciences. Previous studies in other populations have not shown
any such association,[6] and it is possible that this represents a unique
finding, as a result of the characteristics of this population. While H.
pylori positivity rates were not measured in this study, it is interesting
to note that higher education levels have been associated with lower
rates of this infection among HCWs in other parts of the world, and
this could explain this result.[17]
The presence and relative frequency of two major dyspeptic
symptom complexes in this study is revealing. While the previous
literature describes most African patients as suffering from ulcer-like
symptoms, in our study reflux-like symptoms were just as common,
and just as likely to be rated as the major symptom. While the
SF-LDQ has two of four question groups related to reflux symptoms
(heartburn and regurgitation) and this may explain the higher
SF-LDQ scores among subjects with a predominantly reflux-based
symptom complex, the quality-of-life impact measured by SF-NDI
was similar for both groups, as with studies in other populations.[11,14]
In these previous studies, ulcer-like and reflux-like symptom
complexes were associated with moderate responses to acid
suppression with omeprazole, in contrast to those with dysmotilitylike symptoms.[18] H. pylori eradication has also been only modestly
effective in treating dyspeptic symptoms in low-prevalence
populations in the Western world. Whether these data can be
extrapolated to Africa, where H. pylori carriage rates are much higher
and the mind-body interplay around dyspepsia may differ, is unclear.

Study limitations and strengths

We studied HCWs, a relatively well-educated population with

greater than average access to healthcare knowledge and resources.
As such, especially given the protective effect of higher levels
of education apparent in this population, the study is likely to
underestimate rates of dyspepsia in the general Rwandan popu
lation. Other potential reasons for a lower rate of dyspepsia in this
population could include better quality of life, as lower quality of
life has been shown to increase the incidence of new dyspepsia in
a previous study.[19] The study population was based in a provincial

1068

town, and results may not be easily extrapolated to people leading a

traditional village lifestyle.
However, the study being performed in HCWs also gives it some
notable strengths. Firstly and most importantly, HCWs have a
shared linguistic and pathophysiological understanding of clinical
symptoms, which aids the validity of comparison with international
cohorts by reducing the cultural gap. Secondly, most study subjects
were able to understand both English and French, reducing the need
to translate the tools. Thirdly, a relatively small, geographically close
study population aided study logistics and maximised recruitment.
Our study nonetheless has several limitations. The cross-sectional
methodology used made it impossible to determine the direction
of effect in the association between dyspepsia and lower quality
of life. It is possible that dyspepsia causes lower quality of life, but
equally plausible that patients with lower quality of life might be
more susceptible to the onset of dyspeptic symptoms; this complex
interplay needs further study. Definitions of dyspepsia are culturally
and linguistically bounded, and despite our efforts to translate and
validate the SF-LDQ and SF-NDI, it is impossible to know whether
these definitions fill exactly the same linguistic and cultural space
when translated. Our study was of uninvestigated dyspepsia, and
thus the study methodology was unable to distinguish organic and
functional causes of dyspepsia, which might have a different natural
history in African populations, although the quality-of-life impact
of each has been similar in other populations.[6] This should be the
subject of future research.

Conclusion

Dyspepsia is a significant symptom complex among HCWs in this

Rwandan tertiary institution, and has a major quality-of-life impact
in a proportion of these. Reflux-like symptoms were just as common
as ulcer-like symptoms, in contrast to findings in endoscopy-based
studies. HCWs have high levels of uncontrolled symptoms, and rates
may be even higher in the general community because of less access
to healthcare resources.
More resources need to be devoted to the diagnosis and management
of upper abdominal symptoms in Rwanda and other similar resourcepoor areas. Finally, further research in population-based cohorts is
needed to characterise prevalence, symptom-complex patterns and
underlying diagnoses in the wider African population.
Author contributions. TDW conceived the study; RB, BK, PK and TDW
designed the study; PB, RB and JTO carried out and supervised the
interviews; RB, FK and TDW interpreted the data; TDW and RB drafted
the manuscript; and RB, PB, FK, JTO, PK, BK and TDW critically revised
the manuscript for intellectual content. All the authors read and approved
the final manuscript. TDW and RB are guarantors of the paper.
Acknowledgements. We thank Profs Nick Talley and Brendan Delaney
for kind permission to use the SF-NDI and SF-LDQ tools, and Rwanda
Military Hospital, as well as all staff at Butare University Teaching Hospital
for their valued support of this study.
Funding. Rwanda Military Hospital provided study staff, equipment and
facilitation for the conduct of this study. The drafting of this study was
supported in part by Africa Research Initiatives and Support Enterprise
(ARISE), Grant Number W.07.10.102 funded by The Netherlands
African Partnership for Clinical Trials and Capacity Building Program
(NACCAP) of The Netherlands Organisation for Scientific Research
The Netherlands Foundation for the Advancement of Tropical Research
(NWO/WOTRO). Its contents are solely the responsibility of the authors
and do not necessarily represent the official views of the supporting offices.

December 2015, Vol. 105, No. 12

RESEARCH

References
1. Bodger K, Eastwood PG, Manning SI, et al. Dyspepsia workload in urban general practice and
implications of the British Society of Gastroenterology Dyspepsia Guidelines (1996). Aliment
Pharmacol Ther 2000;14(4):413-420.
2. Ford AC, Marwaha A, Sood R, et al. Global prevalence of, and risk factors for, uninvestigated
dyspepsia: A meta-analysis. Gut 2015;64:1049-1057. [http://dx.doi.org/10.1136/gutjnl-2014-307843]
3. Ihezue CH, Oluwole FS, Onuminya JE, et al. Dyspepsias among the highlanders of Nigeria: An
epidemiological survey. Afr J Med Med Sci 1996;25(1):23-29.
4. Holcombe C, Omotara BA, Padonu MK, et al. The prevalence of symptoms of dyspepsia in north
eastern Nigeria: A random community based survey. Trop Geogr Med 1991;43(1-2):209-214.
5. Ford AC, Forman D, Bailey AG, et al. Effect of dyspepsia on survival: A longitudinal 10-year follow-up
study. Am J Gastroenterol 2012;107(6):912-921. [http://dx.doi.org/10.1038/ajg.2012.69]
6. Aro P, Talley NJ, Agreus L, et al. Functional dyspepsia impairs quality of life in the adult
population. Aliment Pharmacol Ther 2011;33(11):1215-1224. [http://dx.doi.org/10.1111/j.13652036.2011.04640.x]
7. Moayyedi P, Duffett S, Braunholtz D, et al. The Leeds Dyspepsia Questionnaire: A valid tool for
measuring the presence and severity of dyspepsia. Aliment Pharmacol Ther 1998;12(12):1257-1262.
8. Fraser A, Delaney BC, Ford AC, et al. The Short-Form Leeds Dyspepsia Questionnaire validation study.
Aliment Pharmacol Ther 2007;25(4):477-486. [http://dx.doi.org/10.1111/j.1365-2036.2006.03233.x]
9. Gatta L, Moayyedi P, Tosetti C, et al. A validation study of the Italian Short-Form Leeds Dyspepsia
Questionnaire. Intern Emerg Med 2010;5(6):501-506. [http://dx.doi.org/10.1007/s11739-010-0467-5]
10. Mahadeva S, Chan WK, Mohazmi M, et al. Validation study of the Leeds Dyspepsia Questionnaire
in a multi-ethnic Asian population. J Gastroenterol Hepatol 2011;26(11):1669-1676. [http://dx.doi.
org/10.1111/j.1440-1746.2011.06806.x]
11. Talley NJ, Verlinden M, Jones M. Quality of life in functional dyspepsia: Responsiveness of the
Nepean Dyspepsia Index and development of a new 10-item short form. Aliment Pharmacol Ther
2001;15(2):207-216.

1069

12. Talley NJ, Verlinden M, Jones M. Validity of a new quality of life scale for functional dyspepsia: A
United States multicenter trial of the Nepean Dyspepsia Index. Am J Gastroenterol 1999;94(9):23902397. [http://dx.doi.org/10.1111/j.1572-0241.1999.01363.x]
13. Tian XP, Li Y, Liang FR, et al. Translation and validation of the Nepean Dyspepsia Index for functional
dyspepsia in China. World J Gastroenterol 2009;15(25):3173-3177.
14. Mahadeva S, Wee HL, Goh KL, et al. Quality of life in South East Asian patients who consult for
dyspepsia: Validation of the short form Nepean Dyspepsia Index. Health Qual Life Outcomes
2009;7:45. [http://dx.doi.org/10.1186/1477-7525-7-45]
15. Bytzer P, Talley NJ. Dyspepsia. Ann Intern Med 2001;134(9):815-822. [http://dx.doi.org/10.7326/00034819-134-9_Part_2-200105011-00004]
16. Mirbagheri SS, Mirbagheri SA, Nabavizadeh B, et al. Impact of microscopic duodenitis on symptomatic
response to Helicobacter pylori eradication in functional dyspepsia. Dig Dis Sci 2015;60(1):163-167.
[http://dx.doi.org/10.1007/s10620-014-3285-1]
17. Triantafillidis JK, Gikas A, Hyphantis T, et al. Helicobacter pylori infection in hospital workers
over a 5-year period: Correlation with demographic and clinical parameters. J Gastroenterol
2002;37(12):1005-1013.
18. Talley NJ, Meineche-Schmidt V, Pare P, et al. Efficacy of omeprazole in functional dyspepsia: Doubleblind, randomized, placebo-controlled trials (the Bond and Opera studies). Aliment Pharmacol Ther
1998;12(11):1055-1065.
19. Ford AC, Forman D, Bailey AG, et al. Initial poor quality of life and new onset of dyspepsia: Results
from a longitudinal 10-year follow-up study. Gut 2007;56(3):321-327. [http://dx.doi.org/10.1136/
gut.2006.099846]

Accepted 5 October 2015.

December 2015, Vol. 105, No. 12